Investigate - European Soccer Database

Table of Contents

• Introduction
• Data Wrangling
• Investigae first question
• Investigae second question
• Investigae thierd question
• Exploratory Data Analysis
• Conclusions

Introduction

Dataset Description

We will use European Soccer Database from Kaggle cover 11 European Leagues from 2008:2016 sessions to analsis the following table displays the dataset in details

Tables names	NO.of rows	NO.of columns
Country	11	2
League	11	3
Player	11.1k	7
Player_Attributes	184k	42
Match	26,000	115
Team	299	5
Team_Attributes	1458	25

• I'm used col player_name from the df "Player" to select mo salah id for merging with df Player_Attributes to explore his performance

• I'm selecting the attributes of only Spanish teams by inner join between the Spanish part of Team_df and Team_Attributes_df

• Spanish teams by inner join between the Spanish part of Team_df and match df

Questions for Analysis

• ### Investigate Mohamed's salah performance and career the ups and down if exists from 2008:2016 as the most Egyptian professional player all over the Egyptian football history we will explore his attributes to find which one increase or decrease in what time to relate it with the teams he played for it which one enhanced his performance and which do opposite.

• ### what are the strongest teams in the Spanish league and the main characters' defense plying style or offensive style? so we will dive the attributes into related styles explore the most important Spanish teams attributes to figure out the top ten strong teams

• ### There is a reputation about the Spanish League that it lacks competition and that the championship is always between Barcelona and Real Madrid, was it the case in the eight years of research? we will explore the Spanish teams goals and total point per season we will rank them and check the claim accuracy

Sir kindly notice that I installed

plotly-5.5.0 tenacity-8.0.1

locally to run two charts below

import numpy as np  
import pandas as pd  
import sqlite3
import matplotlib.pyplot as plt
import matplotlib.patches as patches
import seaborn as sns
import plotly.express as px
from plotly.subplots import make_subplots
import plotly.graph_objects as go 
%matplotlib inline
pd.set_option('display.max_columns', 120)
pd.set_option('display.max_rows', 155)
pd.options.mode.chained_assignment = None  

Data Wrangling

First, let's load all tables from database.sqlite and get familiar with it

connection = sqlite3.connect("database.sqlite")

tables = pd.read_sql("""SELECT *
                        FROM sqlite_master
                        WHERE type='table'
                     """, connection)
tables.head(2)

	type	name	tbl_name	rootpage	sql
0	table	sqlite_sequence	sqlite_sequence	4	CREATE TABLE sqlite_sequence(name,seq)
1	table	Player_Attributes	Player_Attributes	11	CREATE TABLE "Player_Attributes" (\n\t`id`\tIN...

players = pd.read_sql(''' SELECT * 
                          FROM Player
                      ''', connection)
players.head(2)

	id	player_api_id	player_name	player_fifa_api_id	birthday	height	weight
0	1	505942	Aaron Appindangoye	218353	1992-02-29 00:00:00	182.88	187
1	2	155782	Aaron Cresswell	189615	1989-12-15 00:00:00	170.18	146

players.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 11060 entries, 0 to 11059
Data columns (total 7 columns):
 #   Column              Non-Null Count  Dtype  
---  ------              --------------  -----  
 0   id                  11060 non-null  int64  
 1   player_api_id       11060 non-null  int64  
 2   player_name         11060 non-null  object 
 3   player_fifa_api_id  11060 non-null  int64  
 4   birthday            11060 non-null  object 
 5   height              11060 non-null  float64
 6   weight              11060 non-null  int64  
dtypes: float64(1), int64(4), object(2)
memory usage: 605.0+ KB

stats about numeric attributes

players["height"].describe()

count    11060.000000
mean       181.867445
std          6.369201
min        157.480000
25%        177.800000
50%        182.880000
75%        185.420000
max        208.280000
Name: height, dtype: float64

ax=players.boxplot("height")
#ax.set_xlabel("height")
ax.set_ylabel("NO. of Plyers")
ax.set_title("height Box plot")

Text(0.5, 1.0, 'height Box plot')

perfect normal distrubioation

players["weight"].describe()

count    11060.000000
mean       168.380289
std         14.990217
min        117.000000
25%        159.000000
50%        168.000000
75%        179.000000
max        243.000000
Name: weight, dtype: float64

ax=players["weight"].hist(bins=10)
ax.set_xlabel("weight")
ax.set_ylabel("NO. of Plyers")
ax.set_title("weight distrubtion")

Text(0.5, 1.0, 'weight distrubtion')

as we notce that mean and median in both coloumns are close this normal distribution for height and maximum weight is weight 110.223 kg let's explore the heaviest plyer/s and the tallest/s

T=players["height"].sort_values( ).max()

players.query(f"height =={T}")

	id	player_api_id	player_name	player_fifa_api_id	birthday	height	weight
5901	5908	148325	Kristof van Hout	185306	1987-02-09 00:00:00	208.28	243

H=players["weight"].sort_values( ).max()

players.query(f"weight =={H}")

	id	player_api_id	player_name	player_fifa_api_id	birthday	height	weight
5901	5908	148325	Kristof van Hout	185306	1987-02-09 00:00:00	208.28	243
10297	10313	27313	Tim Wiese	53012	1981-12-17 00:00:00	193.04	243

note the weight in pounds

Player_Attributes = pd.read_sql(''' SELECT * 
                          FROM Player_Attributes
                      ''', connection)
Player_Attributes.head(2)

	id	player_fifa_api_id	player_api_id	date	overall_rating	potential	preferred_foot	attacking_work_rate	defensive_work_rate	crossing	finishing	heading_accuracy	short_passing	volleys	dribbling	curve	free_kick_accuracy	long_passing	ball_control	acceleration	sprint_speed	agility	reactions	balance	shot_power	jumping	stamina	strength	long_shots	aggression	interceptions	positioning	vision	penalties	marking	standing_tackle	sliding_tackle	gk_diving	gk_handling	gk_kicking	gk_positioning	gk_reflexes
0	1	218353	505942	2016-02-18 00:00:00	67.0	71.0	right	medium	medium	49.0	44.0	71.0	61.0	44.0	51.0	45.0	39.0	64.0	49.0	60.0	64.0	59.0	47.0	65.0	55.0	58.0	54.0	76.0	35.0	71.0	70.0	45.0	54.0	48.0	65.0	69.0	69.0	6.0	11.0	10.0	8.0	8.0
1	2	218353	505942	2015-11-19 00:00:00	67.0	71.0	right	medium	medium	49.0	44.0	71.0	61.0	44.0	51.0	45.0	39.0	64.0	49.0	60.0	64.0	59.0	47.0	65.0	55.0	58.0	54.0	76.0	35.0	71.0	70.0	45.0	54.0	48.0	65.0	69.0	69.0	6.0	11.0	10.0	8.0	8.0

too meny null but we intrest just in mo salah requred

Player_Attributes.info() 

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 183978 entries, 0 to 183977
Data columns (total 42 columns):
 #   Column               Non-Null Count   Dtype  
---  ------               --------------   -----  
 0   id                   183978 non-null  int64  
 1   player_fifa_api_id   183978 non-null  int64  
 2   player_api_id        183978 non-null  int64  
 3   date                 183978 non-null  object 
 4   overall_rating       183142 non-null  float64
 5   potential            183142 non-null  float64
 6   preferred_foot       183142 non-null  object 
 7   attacking_work_rate  180748 non-null  object 
 8   defensive_work_rate  183142 non-null  object 
 9   crossing             183142 non-null  float64
 10  finishing            183142 non-null  float64
 11  heading_accuracy     183142 non-null  float64
 12  short_passing        183142 non-null  float64
 13  volleys              181265 non-null  float64
 14  dribbling            183142 non-null  float64
 15  curve                181265 non-null  float64
 16  free_kick_accuracy   183142 non-null  float64
 17  long_passing         183142 non-null  float64
 18  ball_control         183142 non-null  float64
 19  acceleration         183142 non-null  float64
 20  sprint_speed         183142 non-null  float64
 21  agility              181265 non-null  float64
 22  reactions            183142 non-null  float64
 23  balance              181265 non-null  float64
 24  shot_power           183142 non-null  float64
 25  jumping              181265 non-null  float64
 26  stamina              183142 non-null  float64
 27  strength             183142 non-null  float64
 28  long_shots           183142 non-null  float64
 29  aggression           183142 non-null  float64
 30  interceptions        183142 non-null  float64
 31  positioning          183142 non-null  float64
 32  vision               181265 non-null  float64
 33  penalties            183142 non-null  float64
 34  marking              183142 non-null  float64
 35  standing_tackle      183142 non-null  float64
 36  sliding_tackle       181265 non-null  float64
 37  gk_diving            183142 non-null  float64
 38  gk_handling          183142 non-null  float64
 39  gk_kicking           183142 non-null  float64
 40  gk_positioning       183142 non-null  float64
 41  gk_reflexes          183142 non-null  float64
dtypes: float64(35), int64(3), object(4)
memory usage: 59.0+ MB

all numeric columns in Player Attributes df is a float which is the appropriate data type

dis_patt = Player_Attributes.iloc[:,np.r_[4,5, 9 :42] ].describe()
dis_patt.iloc[np.r_[1,3,4,5,6,7] , :] 

	overall_rating	potential	crossing	finishing	heading_accuracy	short_passing	volleys	dribbling	curve	free_kick_accuracy	long_passing	ball_control	acceleration	sprint_speed	agility	reactions	balance	shot_power	jumping	stamina	strength	long_shots	aggression	interceptions	positioning	vision	penalties	marking	standing_tackle	sliding_tackle	gk_diving	gk_handling	gk_kicking	gk_positioning	gk_reflexes
mean	68.600015	73.460353	55.086883	49.921078	57.266023	62.429672	49.468436	59.175154	52.965675	49.38095	57.06988	63.388879	67.659357	68.051244	65.97091	66.103706	65.189496	61.808427	66.969045	67.038544	67.424529	53.339431	60.948046	52.009271	55.786504	57.87355	55.003986	46.772242	50.351257	48.001462	14.704393	16.063612	20.998362	16.132154	16.441439
min	33.000000	39.000000	1.000000	1.000000	1.000000	3.000000	1.000000	1.000000	2.000000	1.00000	3.00000	5.000000	10.000000	12.000000	11.00000	17.000000	12.000000	2.000000	14.000000	10.000000	10.000000	1.000000	6.000000	1.000000	2.000000	1.00000	2.000000	1.000000	1.000000	2.000000	1.000000	1.000000	1.000000	1.000000	1.000000
25%	64.000000	69.000000	45.000000	34.000000	49.000000	57.000000	35.000000	52.000000	41.000000	36.00000	49.00000	58.000000	61.000000	62.000000	58.00000	61.000000	58.000000	54.000000	60.000000	61.000000	60.000000	41.000000	51.000000	34.000000	45.000000	49.00000	45.000000	25.000000	29.000000	25.000000	7.000000	8.000000	8.000000	8.000000	8.000000
50%	69.000000	74.000000	59.000000	53.000000	60.000000	65.000000	52.000000	64.000000	56.000000	50.00000	59.00000	67.000000	69.000000	69.000000	68.00000	67.000000	67.000000	65.000000	68.000000	69.000000	69.000000	58.000000	64.000000	57.000000	60.000000	60.00000	57.000000	50.000000	56.000000	53.000000	10.000000	11.000000	12.000000	11.000000	11.000000
75%	73.000000	78.000000	68.000000	65.000000	68.000000	72.000000	64.000000	72.000000	67.000000	63.00000	67.00000	73.000000	77.000000	77.000000	75.00000	72.000000	74.000000	73.000000	74.000000	76.000000	76.000000	67.000000	73.000000	68.000000	69.000000	69.00000	67.000000	66.000000	69.000000	67.000000	13.000000	15.000000	15.000000	15.000000	15.000000
max	94.000000	97.000000	95.000000	97.000000	98.000000	97.000000	93.000000	97.000000	94.000000	97.00000	97.00000	97.000000	97.000000	97.000000	96.00000	96.000000	96.000000	97.000000	96.000000	96.000000	96.000000	96.000000	97.000000	96.000000	96.000000	97.00000	96.000000	96.000000	95.000000	95.000000	94.000000	93.000000	97.000000	96.000000	96.000000

Team_df = pd.read_sql(''' SELECT * 
                          FROM Team
                      ''', connection)
Team_df.head(2)

	id	team_api_id	team_fifa_api_id	team_long_name	team_short_name
0	1	9987	673.0	KRC Genk	GEN
1	2	9993	675.0	Beerschot AC	BAC

Team_df.info() #team_fifa_api_id null

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 299 entries, 0 to 298
Data columns (total 5 columns):
 #   Column            Non-Null Count  Dtype  
---  ------            --------------  -----  
 0   id                299 non-null    int64  
 1   team_api_id       299 non-null    int64  
 2   team_fifa_api_id  288 non-null    float64
 3   team_long_name    299 non-null    object 
 4   team_short_name   299 non-null    object 
dtypes: float64(1), int64(2), object(2)
memory usage: 11.8+ KB

Team_Attributes_df = pd.read_sql(''' SELECT * 
                          FROM Team_Attributes
                      ''', connection)
Team_Attributes_df.head(2) 

	id	team_fifa_api_id	team_api_id	date	buildUpPlaySpeed	buildUpPlaySpeedClass	buildUpPlayDribbling	buildUpPlayDribblingClass	buildUpPlayPassing	buildUpPlayPassingClass	buildUpPlayPositioningClass	chanceCreationPassing	chanceCreationPassingClass	chanceCreationCrossing	chanceCreationCrossingClass	chanceCreationShooting	chanceCreationShootingClass	chanceCreationPositioningClass	defencePressure	defencePressureClass	defenceAggression	defenceAggressionClass	defenceTeamWidth	defenceTeamWidthClass	defenceDefenderLineClass
0	1	434	9930	2010-02-22 00:00:00	60	Balanced	NaN	Little	50	Mixed	Organised	60	Normal	65	Normal	55	Normal	Organised	50	Medium	55	Press	45	Normal	Cover
1	2	434	9930	2014-09-19 00:00:00	52	Balanced	48.0	Normal	56	Mixed	Organised	54	Normal	63	Normal	64	Normal	Organised	47	Medium	44	Press	54	Normal	Cover

Team_Attributes_df.info()  #buildUpPlayDribbling null

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1458 entries, 0 to 1457
Data columns (total 25 columns):
 #   Column                          Non-Null Count  Dtype  
---  ------                          --------------  -----  
 0   id                              1458 non-null   int64  
 1   team_fifa_api_id                1458 non-null   int64  
 2   team_api_id                     1458 non-null   int64  
 3   date                            1458 non-null   object 
 4   buildUpPlaySpeed                1458 non-null   int64  
 5   buildUpPlaySpeedClass           1458 non-null   object 
 6   buildUpPlayDribbling            489 non-null    float64
 7   buildUpPlayDribblingClass       1458 non-null   object 
 8   buildUpPlayPassing              1458 non-null   int64  
 9   buildUpPlayPassingClass         1458 non-null   object 
 10  buildUpPlayPositioningClass     1458 non-null   object 
 11  chanceCreationPassing           1458 non-null   int64  
 12  chanceCreationPassingClass      1458 non-null   object 
 13  chanceCreationCrossing          1458 non-null   int64  
 14  chanceCreationCrossingClass     1458 non-null   object 
 15  chanceCreationShooting          1458 non-null   int64  
 16  chanceCreationShootingClass     1458 non-null   object 
 17  chanceCreationPositioningClass  1458 non-null   object 
 18  defencePressure                 1458 non-null   int64  
 19  defencePressureClass            1458 non-null   object 
 20  defenceAggression               1458 non-null   int64  
 21  defenceAggressionClass          1458 non-null   object 
 22  defenceTeamWidth                1458 non-null   int64  
 23  defenceTeamWidthClass           1458 non-null   object 
 24  defenceDefenderLineClass        1458 non-null   object 
dtypes: float64(1), int64(11), object(13)
memory usage: 284.9+ KB

Team_Attributes_df.iloc[: ,np.r_ [4,7,10,12,14,17,19,21]]

	buildUpPlaySpeed	buildUpPlayDribblingClass	buildUpPlayPositioningClass	chanceCreationPassingClass	chanceCreationCrossingClass	chanceCreationPositioningClass	defencePressureClass	defenceAggressionClass
0	60	Little	Organised	Normal	Normal	Organised	Medium	Press
1	52	Normal	Organised	Normal	Normal	Organised	Medium	Press
2	47	Normal	Organised	Normal	Normal	Organised	Medium	Press
3	70	Little	Organised	Risky	Lots	Organised	Medium	Double
4	47	Little	Organised	Normal	Normal	Organised	Medium	Press
...	...	...	...	...	...	...	...	...
1453	52	Little	Organised	Normal	Normal	Organised	Medium	Press
1454	54	Little	Organised	Normal	Normal	Organised	Medium	Press
1455	54	Little	Organised	Normal	Normal	Organised	Medium	Press
1456	54	Normal	Organised	Normal	Normal	Organised	Medium	Press
1457	54	Normal	Organised	Normal	Normal	Organised	Medium	Press

1458 rows × 8 columns

match_df = pd.read_sql(''' SELECT * 
                          FROM Match
                      ''', connection)
match_df.head(2)

	id	country_id	league_id	season	stage	date	match_api_id	home_team_api_id	away_team_api_id	home_team_goal	away_team_goal	home_player_X1	home_player_X2	home_player_X3	home_player_X4	home_player_X5	home_player_X6	home_player_X7	home_player_X8	home_player_X9	home_player_X10	home_player_X11	away_player_X1	away_player_X2	away_player_X3	away_player_X4	away_player_X5	away_player_X6	away_player_X7	away_player_X8	away_player_X9	away_player_X10	away_player_X11	home_player_Y1	home_player_Y2	home_player_Y3	home_player_Y4	home_player_Y5	home_player_Y6	home_player_Y7	home_player_Y8	home_player_Y9	home_player_Y10	home_player_Y11	away_player_Y1	away_player_Y2	away_player_Y3	away_player_Y4	away_player_Y5	away_player_Y6	away_player_Y7	away_player_Y8	away_player_Y9	away_player_Y10	away_player_Y11	home_player_1	home_player_2	home_player_3	home_player_4	home_player_5	home_player_6	home_player_7	home_player_8	home_player_9	home_player_10	home_player_11	away_player_1	away_player_2	away_player_3	away_player_4	away_player_5	away_player_6	away_player_7	away_player_8	away_player_9	away_player_10	away_player_11	goal	shoton	shotoff	foulcommit	card	cross	corner	possession	B365H	B365D	B365A	BWH	BWD	BWA	IWH	IWD	IWA	LBH	LBD	LBA	PSH	PSD	PSA	WHH	WHD	WHA	SJH	SJD	SJA	VCH	VCD	VCA	GBH	GBD	GBA	BSH	BSD	BSA
0	1	1	1	2008/2009	1	2008-08-17 00:00:00	492473	9987	9993	1	1	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	None	None	None	None	None	None	None	None	1.73	3.4	5.0	1.75	3.35	4.20	1.85	3.2	3.5	1.8	3.3	3.75	NaN	NaN	NaN	1.70	3.3	4.33	1.90	3.3	4.0	1.65	3.40	4.50	1.78	3.25	4.00	1.73	3.40	4.2
1	2	1	1	2008/2009	1	2008-08-16 00:00:00	492474	10000	9994	0	0	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	None	None	None	None	None	None	None	None	1.95	3.2	3.6	1.80	3.30	3.95	1.90	3.2	3.5	1.9	3.2	3.50	NaN	NaN	NaN	1.83	3.3	3.60	1.95	3.3	3.8	2.00	3.25	3.25	1.85	3.25	3.75	1.91	3.25	3.6

match_df.info(verbose=True, show_counts=True  ) # too meny null bute we will not use them
#match_df.iloc[:,np.r_[5,3, 9,10 ] ]

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 25979 entries, 0 to 25978
Data columns (total 115 columns):
 #    Column            Non-Null Count  Dtype  
---   ------            --------------  -----  
 0    id                25979 non-null  int64  
 1    country_id        25979 non-null  int64  
 2    league_id         25979 non-null  int64  
 3    season            25979 non-null  object 
 4    stage             25979 non-null  int64  
 5    date              25979 non-null  object 
 6    match_api_id      25979 non-null  int64  
 7    home_team_api_id  25979 non-null  int64  
 8    away_team_api_id  25979 non-null  int64  
 9    home_team_goal    25979 non-null  int64  
 10   away_team_goal    25979 non-null  int64  
 11   home_player_X1    24158 non-null  float64
 12   home_player_X2    24158 non-null  float64
 13   home_player_X3    24147 non-null  float64
 14   home_player_X4    24147 non-null  float64
 15   home_player_X5    24147 non-null  float64
 16   home_player_X6    24147 non-null  float64
 17   home_player_X7    24147 non-null  float64
 18   home_player_X8    24147 non-null  float64
 19   home_player_X9    24147 non-null  float64
 20   home_player_X10   24147 non-null  float64
 21   home_player_X11   24147 non-null  float64
 22   away_player_X1    24147 non-null  float64
 23   away_player_X2    24147 non-null  float64
 24   away_player_X3    24147 non-null  float64
 25   away_player_X4    24147 non-null  float64
 26   away_player_X5    24147 non-null  float64
 27   away_player_X6    24147 non-null  float64
 28   away_player_X7    24147 non-null  float64
 29   away_player_X8    24147 non-null  float64
 30   away_player_X9    24146 non-null  float64
 31   away_player_X10   24146 non-null  float64
 32   away_player_X11   24140 non-null  float64
 33   home_player_Y1    24158 non-null  float64
 34   home_player_Y2    24158 non-null  float64
 35   home_player_Y3    24147 non-null  float64
 36   home_player_Y4    24147 non-null  float64
 37   home_player_Y5    24147 non-null  float64
 38   home_player_Y6    24147 non-null  float64
 39   home_player_Y7    24147 non-null  float64
 40   home_player_Y8    24147 non-null  float64
 41   home_player_Y9    24147 non-null  float64
 42   home_player_Y10   24147 non-null  float64
 43   home_player_Y11   24147 non-null  float64
 44   away_player_Y1    24147 non-null  float64
 45   away_player_Y2    24147 non-null  float64
 46   away_player_Y3    24147 non-null  float64
 47   away_player_Y4    24147 non-null  float64
 48   away_player_Y5    24147 non-null  float64
 49   away_player_Y6    24147 non-null  float64
 50   away_player_Y7    24147 non-null  float64
 51   away_player_Y8    24147 non-null  float64
 52   away_player_Y9    24146 non-null  float64
 53   away_player_Y10   24146 non-null  float64
 54   away_player_Y11   24140 non-null  float64
 55   home_player_1     24755 non-null  float64
 56   home_player_2     24664 non-null  float64
 57   home_player_3     24698 non-null  float64
 58   home_player_4     24656 non-null  float64
 59   home_player_5     24663 non-null  float64
 60   home_player_6     24654 non-null  float64
 61   home_player_7     24752 non-null  float64
 62   home_player_8     24670 non-null  float64
 63   home_player_9     24706 non-null  float64
 64   home_player_10    24543 non-null  float64
 65   home_player_11    24424 non-null  float64
 66   away_player_1     24745 non-null  float64
 67   away_player_2     24701 non-null  float64
 68   away_player_3     24686 non-null  float64
 69   away_player_4     24658 non-null  float64
 70   away_player_5     24644 non-null  float64
 71   away_player_6     24666 non-null  float64
 72   away_player_7     24744 non-null  float64
 73   away_player_8     24638 non-null  float64
 74   away_player_9     24651 non-null  float64
 75   away_player_10    24538 non-null  float64
 76   away_player_11    24425 non-null  float64
 77   goal              14217 non-null  object 
 78   shoton            14217 non-null  object 
 79   shotoff           14217 non-null  object 
 80   foulcommit        14217 non-null  object 
 81   card              14217 non-null  object 
 82   cross             14217 non-null  object 
 83   corner            14217 non-null  object 
 84   possession        14217 non-null  object 
 85   B365H             22592 non-null  float64
 86   B365D             22592 non-null  float64
 87   B365A             22592 non-null  float64
 88   BWH               22575 non-null  float64
 89   BWD               22575 non-null  float64
 90   BWA               22575 non-null  float64
 91   IWH               22520 non-null  float64
 92   IWD               22520 non-null  float64
 93   IWA               22520 non-null  float64
 94   LBH               22556 non-null  float64
 95   LBD               22556 non-null  float64
 96   LBA               22556 non-null  float64
 97   PSH               11168 non-null  float64
 98   PSD               11168 non-null  float64
 99   PSA               11168 non-null  float64
 100  WHH               22571 non-null  float64
 101  WHD               22571 non-null  float64
 102  WHA               22571 non-null  float64
 103  SJH               17097 non-null  float64
 104  SJD               17097 non-null  float64
 105  SJA               17097 non-null  float64
 106  VCH               22568 non-null  float64
 107  VCD               22568 non-null  float64
 108  VCA               22568 non-null  float64
 109  GBH               14162 non-null  float64
 110  GBD               14162 non-null  float64
 111  GBA               14162 non-null  float64
 112  BSH               14161 non-null  float64
 113  BSD               14161 non-null  float64
 114  BSA               14161 non-null  float64
dtypes: float64(96), int64(9), object(10)
memory usage: 22.8+ MB

we ignored buildUpPlayDribbling will droped it later

 match_df.iloc[:,np.r_[5,3, 9,10 ] ].describe()

	home_team_goal	away_team_goal
count	25979.000000	25979.000000
mean	1.544594	1.160938
std	1.297158	1.142110
min	0.000000	0.000000
25%	1.000000	0.000000
50%	1.000000	1.000000
75%	2.000000	2.000000
max	10.000000	9.000000

match_df["home_team_goal"].hist(bins=9);
# ax=players["weight"].hist()
# ax.set_xlabel("weight")
# ax.set_ylabel("NO. of Plyers")
# ax.set_title("weight distrubtion")

match_df["away_team_goal"].hist(bins=9);

of course, the 2 charts show strong right-skewed which more values are concentrated on the lift side of the distribution graph while the ri9ght tail of the distribution graph is longer in our case and for any football ((soccer)) fan it's normal that almost matches end with a few goals or even without a goal at all min =0

let's drop all columns which 40% of them had a nulls value

#define the function 
     
def drop_forty_null_column(df):
    """ drop any columns contain 40% or more  of null values  
        Args:  df - pandas DataFrame
        return  the modified df  - pandas DataFrame
     """
    
    col_list = df.columns
    while True:
        for col in col_list :
            if (df[col].isnull().sum()/len(df)) >= 0.4 :

                df.drop(columns=[col],inplace =True )
            else :
                continue


        return df

Team_Attributes_df.shape

(1458, 25)

drop_forty_null_column(Team_Attributes_df)
Team_Attributes_df.shape   # drop buildUpPlayDribbling col 

(1458, 24)

match_df.shape

(25979, 115)

drop_forty_null_column(match_df)
match_df.shape   # alots of cols in this DF  have 40%  or more of null values

(25979, 98)

---

First question

Data Cleaning

merge player and player_att to creat a new df contains the player names and attributes

Player_Att_numric=Player_Attributes.iloc[:,np.r_[2,3,4,5, 9 :42] ]
merged_plyerName_plyerAtt = players.merge(Player_Att_numric, on="player_api_id",suffixes =("_name","_att"))

select mohamed salah records by mask

let's chek mo' salah df for nulls values

mo_salah_df=merged_plyerName_plyerAtt[merged_plyerName_plyerAtt["player_name"]== "Mohamed Salah"]
# lets chik if  any null values !
mo_salah_df.isnull().values.any()
mo_salah_df.duplicated().sum()

0

mo_salah_df.duplicated().sum()

0

Great no null values or dupliate values found

# let's have adeepr look of mo's atributes 
mo_salah_explor_df =mo_salah_df.iloc[: , np.r_ [8:40 ]]

Exploratory Data Analysis for first question

plt.figure(figsize=(12,15), dpi= 180)
mo_mean=mo_salah_df.iloc[:,np.r_[ 9 :33] ].mean() 
total_mean =Player_Attributes.iloc[:,np.r_[ 9 :34] ].mean()
fig, ax = plt.subplots(figsize =(8,6))
ax.hist(total_mean, alpha=0.5, label='genrral mean', bins =7)
ax.hist(mo_mean, alpha=0.5, label='mo_salah mean',bins =7)
ax.set_title("Mo Salah attributes average vs total players attributes average")
ax.set_xlabel('Performance')
ax.set_ylabel('Attributes No.')
ax.legend(loc='upper right')
ax.legend(bbox_to_anchor=(1,1))
plt.show();

<Figure size 2160x2700 with 0 Axes>

The mo's average is clearly much bigger than the main average

# let's have adeepr look of mo's atributes 
mo_salah_explor_df =mo_salah_df.iloc[: , np.r_ [7:33 ]]

mo_salah_explor_df

	date	overall_rating	potential	crossing	finishing	heading_accuracy	short_passing	volleys	dribbling	curve	free_kick_accuracy	long_passing	ball_control	acceleration	sprint_speed	agility	reactions	balance	shot_power	jumping	stamina	strength	long_shots	aggression	interceptions	positioning
128023	2016-05-05 00:00:00	80.0	85.0	68.0	77.0	59.0	75.0	69.0	86.0	59.0	60.0	63.0	86.0	92.0	91.0	86.0	80.0	84.0	71.0	70.0	77.0	60.0	61.0	63.0	55.0	77.0
128024	2016-04-21 00:00:00	80.0	85.0	68.0	77.0	59.0	75.0	69.0	86.0	59.0	60.0	63.0	86.0	92.0	91.0	86.0	75.0	84.0	75.0	70.0	77.0	60.0	61.0	63.0	55.0	77.0
128025	2015-12-17 00:00:00	80.0	85.0	68.0	77.0	59.0	75.0	69.0	86.0	64.0	68.0	67.0	86.0	92.0	91.0	86.0	75.0	84.0	75.0	70.0	77.0	60.0	61.0	63.0	55.0	77.0
128026	2015-11-12 00:00:00	80.0	85.0	68.0	77.0	59.0	75.0	69.0	86.0	64.0	68.0	67.0	86.0	92.0	91.0	86.0	75.0	84.0	75.0	70.0	77.0	60.0	61.0	63.0	55.0	77.0
128027	2015-10-09 00:00:00	80.0	87.0	68.0	77.0	59.0	75.0	69.0	86.0	64.0	68.0	67.0	86.0	92.0	91.0	86.0	75.0	84.0	75.0	70.0	77.0	60.0	61.0	63.0	55.0	77.0
128028	2015-09-21 00:00:00	80.0	87.0	68.0	77.0	59.0	75.0	69.0	86.0	64.0	68.0	67.0	86.0	92.0	91.0	86.0	75.0	84.0	75.0	70.0	77.0	60.0	61.0	63.0	55.0	77.0
128029	2015-07-03 00:00:00	77.0	84.0	67.0	76.0	58.0	74.0	68.0	85.0	63.0	67.0	66.0	85.0	92.0	94.0	86.0	74.0	84.0	74.0	70.0	77.0	60.0	60.0	62.0	54.0	76.0
128030	2015-04-24 00:00:00	77.0	84.0	67.0	76.0	58.0	74.0	68.0	85.0	63.0	67.0	66.0	85.0	92.0	94.0	86.0	74.0	84.0	74.0	70.0	77.0	60.0	60.0	62.0	54.0	76.0
128031	2015-03-20 00:00:00	76.0	84.0	67.0	74.0	58.0	74.0	68.0	85.0	63.0	67.0	66.0	81.0	92.0	94.0	86.0	70.0	84.0	74.0	70.0	77.0	60.0	50.0	62.0	58.0	74.0
128032	2015-03-13 00:00:00	76.0	84.0	67.0	74.0	58.0	74.0	68.0	85.0	63.0	67.0	66.0	81.0	92.0	94.0	86.0	70.0	84.0	74.0	70.0	77.0	60.0	50.0	62.0	58.0	74.0
128033	2015-03-06 00:00:00	76.0	84.0	67.0	68.0	58.0	74.0	60.0	85.0	63.0	67.0	66.0	81.0	92.0	94.0	86.0	70.0	80.0	74.0	70.0	77.0	59.0	50.0	62.0	58.0	74.0
128034	2015-02-06 00:00:00	76.0	84.0	67.0	68.0	58.0	74.0	48.0	85.0	63.0	67.0	66.0	81.0	92.0	94.0	86.0	69.0	80.0	74.0	70.0	77.0	59.0	50.0	61.0	59.0	73.0
128035	2014-09-18 00:00:00	76.0	84.0	67.0	68.0	58.0	74.0	48.0	85.0	63.0	67.0	66.0	81.0	92.0	94.0	86.0	69.0	80.0	74.0	70.0	77.0	59.0	50.0	61.0	59.0	73.0
128036	2014-05-16 00:00:00	77.0	87.0	65.0	68.0	58.0	74.0	48.0	86.0	63.0	67.0	66.0	89.0	92.0	94.0	86.0	68.0	80.0	74.0	70.0	77.0	58.0	50.0	61.0	59.0	73.0
128037	2014-01-31 00:00:00	77.0	87.0	65.0	68.0	58.0	74.0	48.0	86.0	63.0	67.0	66.0	89.0	92.0	94.0	86.0	68.0	80.0	74.0	70.0	77.0	58.0	50.0	61.0	59.0	73.0
128038	2014-01-10 00:00:00	75.0	86.0	61.0	68.0	58.0	74.0	48.0	82.0	63.0	67.0	61.0	87.0	92.0	93.0	78.0	68.0	71.0	68.0	70.0	77.0	58.0	50.0	61.0	59.0	73.0
128039	2013-12-13 00:00:00	75.0	85.0	61.0	68.0	58.0	74.0	48.0	82.0	63.0	67.0	61.0	87.0	92.0	91.0	78.0	68.0	71.0	68.0	70.0	77.0	58.0	50.0	61.0	59.0	73.0
128040	2013-11-01 00:00:00	75.0	85.0	61.0	68.0	48.0	74.0	48.0	82.0	63.0	67.0	61.0	87.0	92.0	91.0	78.0	68.0	71.0	68.0	70.0	77.0	58.0	50.0	61.0	59.0	73.0
128041	2013-10-04 00:00:00	75.0	85.0	61.0	68.0	48.0	74.0	48.0	82.0	63.0	67.0	61.0	87.0	92.0	91.0	73.0	68.0	60.0	68.0	70.0	77.0	56.0	50.0	61.0	59.0	73.0
128042	2013-09-20 00:00:00	74.0	85.0	61.0	67.0	48.0	74.0	48.0	82.0	63.0	67.0	60.0	86.0	92.0	91.0	73.0	68.0	54.0	68.0	70.0	77.0	56.0	50.0	61.0	59.0	73.0
128043	2013-03-22 00:00:00	73.0	82.0	61.0	66.0	48.0	74.0	46.0	82.0	56.0	59.0	60.0	79.0	89.0	85.0	73.0	68.0	54.0	66.0	70.0	72.0	56.0	51.0	61.0	59.0	71.0
128044	2013-03-08 00:00:00	73.0	82.0	61.0	66.0	48.0	74.0	46.0	82.0	56.0	59.0	60.0	79.0	89.0	85.0	73.0	68.0	54.0	66.0	70.0	72.0	56.0	51.0	61.0	59.0	71.0
128045	2013-02-15 00:00:00	70.0	82.0	61.0	66.0	48.0	74.0	46.0	79.0	56.0	59.0	60.0	68.0	75.0	81.0	73.0	68.0	54.0	66.0	70.0	72.0	56.0	51.0	61.0	59.0	71.0
128046	2012-08-31 00:00:00	70.0	82.0	61.0	66.0	48.0	74.0	46.0	79.0	56.0	59.0	60.0	68.0	75.0	81.0	73.0	68.0	54.0	66.0	70.0	72.0	56.0	51.0	61.0	59.0	73.0
128047	2007-02-22 00:00:00	70.0	82.0	61.0	66.0	48.0	74.0	46.0	79.0	56.0	59.0	60.0	68.0	75.0	81.0	73.0	68.0	54.0	66.0	70.0	72.0	56.0	51.0	61.0	59.0	73.0

mo_salah_explor_df.describe()

	overall_rating	potential	crossing	finishing	heading_accuracy	short_passing	volleys	dribbling	curve	free_kick_accuracy	long_passing	ball_control	acceleration	sprint_speed	agility	reactions	balance	shot_power	jumping	stamina	strength	long_shots	aggression	interceptions	positioning
count	25.000000	25.000000	25.000000	25.000000	25.000000	25.00000	25.000000	25.000000	25.000000	25.000000	25.000000	25.00000	25.000000	25.000000	25.000000	25.000000	25.000000	25.000000	25.0	25.000000	25.000000	25.000	25.000000	25.000000	25.000000
mean	75.920000	84.480000	64.680000	70.840000	55.040000	74.24000	56.320000	83.760000	61.440000	65.000000	63.680000	82.60000	89.720000	90.480000	81.400000	70.680000	73.480000	71.280000	70.0	76.000000	58.360000	53.640	61.680000	57.520000	74.080000
std	3.121431	1.610383	3.158586	4.588754	4.945368	0.43589	10.566141	2.420055	3.028751	3.662877	2.911472	6.22495	5.608921	4.331282	5.937171	3.508561	12.576831	3.702702	0.0	2.041241	1.680278	4.999	0.852447	1.981582	2.039608
min	70.000000	82.000000	61.000000	66.000000	48.000000	74.00000	46.000000	79.000000	56.000000	59.000000	60.000000	68.00000	75.000000	81.000000	73.000000	68.000000	54.000000	66.000000	70.0	72.000000	56.000000	50.000	61.000000	54.000000	71.000000
25%	75.000000	84.000000	61.000000	68.000000	48.000000	74.00000	48.000000	82.000000	59.000000	60.000000	61.000000	81.00000	92.000000	91.000000	73.000000	68.000000	60.000000	68.000000	70.0	77.000000	56.000000	50.000	61.000000	55.000000	73.000000
50%	76.000000	85.000000	67.000000	68.000000	58.000000	74.00000	48.000000	85.000000	63.000000	67.000000	66.000000	86.00000	92.000000	91.000000	86.000000	69.000000	80.000000	74.000000	70.0	77.000000	59.000000	51.000	61.000000	59.000000	73.000000
75%	77.000000	85.000000	67.000000	76.000000	58.000000	74.00000	68.000000	86.000000	63.000000	67.000000	66.000000	86.00000	92.000000	94.000000	86.000000	74.000000	84.000000	74.000000	70.0	77.000000	60.000000	60.000	62.000000	59.000000	76.000000
max	80.000000	87.000000	68.000000	77.000000	59.000000	75.00000	69.000000	86.000000	64.000000	68.000000	67.000000	89.00000	92.000000	94.000000	86.000000	80.000000	84.000000	75.000000	70.0	77.000000	60.000000	61.000	63.000000	59.000000	77.000000

from the recoreds we can find tremendous improvements in mo's performance max-min the important attributes which had a strong impact on his overall rating are :- finishing , heading_accuracy , volleys , ball_control , acceleration ,sprint_speed , agility , reactions , balance

mo_df =mo_salah_df.iloc[: , np.r_[7,8,11,12,14,19,20,21,22,23,24]] 
mo_df.head(34)

	date	overall_rating	finishing	heading_accuracy	volleys	ball_control	acceleration	sprint_speed	agility	reactions	balance
128023	2016-05-05 00:00:00	80.0	77.0	59.0	69.0	86.0	92.0	91.0	86.0	80.0	84.0
128024	2016-04-21 00:00:00	80.0	77.0	59.0	69.0	86.0	92.0	91.0	86.0	75.0	84.0
128025	2015-12-17 00:00:00	80.0	77.0	59.0	69.0	86.0	92.0	91.0	86.0	75.0	84.0
128026	2015-11-12 00:00:00	80.0	77.0	59.0	69.0	86.0	92.0	91.0	86.0	75.0	84.0
128027	2015-10-09 00:00:00	80.0	77.0	59.0	69.0	86.0	92.0	91.0	86.0	75.0	84.0
128028	2015-09-21 00:00:00	80.0	77.0	59.0	69.0	86.0	92.0	91.0	86.0	75.0	84.0
128029	2015-07-03 00:00:00	77.0	76.0	58.0	68.0	85.0	92.0	94.0	86.0	74.0	84.0
128030	2015-04-24 00:00:00	77.0	76.0	58.0	68.0	85.0	92.0	94.0	86.0	74.0	84.0
128031	2015-03-20 00:00:00	76.0	74.0	58.0	68.0	81.0	92.0	94.0	86.0	70.0	84.0
128032	2015-03-13 00:00:00	76.0	74.0	58.0	68.0	81.0	92.0	94.0	86.0	70.0	84.0
128033	2015-03-06 00:00:00	76.0	68.0	58.0	60.0	81.0	92.0	94.0	86.0	70.0	80.0
128034	2015-02-06 00:00:00	76.0	68.0	58.0	48.0	81.0	92.0	94.0	86.0	69.0	80.0
128035	2014-09-18 00:00:00	76.0	68.0	58.0	48.0	81.0	92.0	94.0	86.0	69.0	80.0
128036	2014-05-16 00:00:00	77.0	68.0	58.0	48.0	89.0	92.0	94.0	86.0	68.0	80.0
128037	2014-01-31 00:00:00	77.0	68.0	58.0	48.0	89.0	92.0	94.0	86.0	68.0	80.0
128038	2014-01-10 00:00:00	75.0	68.0	58.0	48.0	87.0	92.0	93.0	78.0	68.0	71.0
128039	2013-12-13 00:00:00	75.0	68.0	58.0	48.0	87.0	92.0	91.0	78.0	68.0	71.0
128040	2013-11-01 00:00:00	75.0	68.0	48.0	48.0	87.0	92.0	91.0	78.0	68.0	71.0
128041	2013-10-04 00:00:00	75.0	68.0	48.0	48.0	87.0	92.0	91.0	73.0	68.0	60.0
128042	2013-09-20 00:00:00	74.0	67.0	48.0	48.0	86.0	92.0	91.0	73.0	68.0	54.0
128043	2013-03-22 00:00:00	73.0	66.0	48.0	46.0	79.0	89.0	85.0	73.0	68.0	54.0
128044	2013-03-08 00:00:00	73.0	66.0	48.0	46.0	79.0	89.0	85.0	73.0	68.0	54.0
128045	2013-02-15 00:00:00	70.0	66.0	48.0	46.0	68.0	75.0	81.0	73.0	68.0	54.0
128046	2012-08-31 00:00:00	70.0	66.0	48.0	46.0	68.0	75.0	81.0	73.0	68.0	54.0
128047	2007-02-22 00:00:00	70.0	66.0	48.0	46.0	68.0	75.0	81.0	73.0	68.0	54.0

mo_df.describe()

	overall_rating	finishing	heading_accuracy	volleys	ball_control	acceleration	sprint_speed	agility	reactions	balance
count	25.000000	25.000000	25.000000	25.000000	25.00000	25.000000	25.000000	25.000000	25.000000	25.000000
mean	75.920000	70.840000	55.040000	56.320000	82.60000	89.720000	90.480000	81.400000	70.680000	73.480000
std	3.121431	4.588754	4.945368	10.566141	6.22495	5.608921	4.331282	5.937171	3.508561	12.576831
min	70.000000	66.000000	48.000000	46.000000	68.00000	75.000000	81.000000	73.000000	68.000000	54.000000
25%	75.000000	68.000000	48.000000	48.000000	81.00000	92.000000	91.000000	73.000000	68.000000	60.000000
50%	76.000000	68.000000	58.000000	48.000000	86.00000	92.000000	91.000000	86.000000	69.000000	80.000000
75%	77.000000	76.000000	58.000000	68.000000	86.00000	92.000000	94.000000	86.000000	74.000000	84.000000
max	80.000000	77.000000	59.000000	69.000000	89.00000	92.000000	94.000000	86.000000	80.000000	84.000000

as we notice that all atributs's

median is > mean

which indecate to left SKEWED negative

let's explore a few single attributes alone to be sure from our conclusion

plt.figure(figsize=(6,6), dpi= 180) 
fig, ax = plt.subplots();
mo_df.overall_rating.hist(bins=4);
ax.set_title("Mo Salah overall_rating")
ax.set_ylabel('frequency')
ax.set_xlabel('overall rating grade');

<Figure size 1080x1080 with 0 Axes>

As we know left-skewed type of distribution in which more values are concentrated on the right (tail) side of the distribution graph while the left tail of the distribution graph is longer in our case indicate to very good improvement mo" Salah performance

plt.figure(figsize=(6,6), dpi= 180) 
fig, ax = plt.subplots();
mo_df.sprint_speed.hist(bins=3);
ax.set_title("Mo Salah sprint_speed")
ax.set_ylabel('frequency')
ax.set_xlabel('sprint speed grade');

<Figure size 1080x1080 with 0 Axes>

almost over 90 % which reflect one of main mo's character as aplyer

plt.figure(figsize=(6,6), dpi= 180) 
fig, ax = plt.subplots();
mo_df.balance.hist(bins=3);
ax.set_title("Mo Salah balance")
ax.set_ylabel('frequency')
ax.set_xlabel('balance grade');

<Figure size 1080x1080 with 0 Axes>

parsing the date column from string to date

then use it to sort mo's records ascendingly

mo_df["date"]=pd.to_datetime(mo_df["date"])
mo_df =mo_df.sort_values(by=['date'], ascending=True)

plt.rcParams.update({'font.size': 19})
cols_names =mo_df.columns[1:]
x = np.arange(len(cols_names))   
bar_size = 0.45    
#plot the chart
fig, ax = plt.subplots(figsize =(15,12), dpi= 180)
ax.barh(x - bar_size/2,mo_df.min().tolist()[1:], bar_size, label= " Mo  Salah minmum  " )
ax.barh(x + bar_size/2, mo_df.max().tolist()[1:], bar_size, label='" Mo  Salah Maxmum  "')

# formating the chart.
ax.set_title("Mo Salah max vs Mo Salah min")
ax.set_ylabel('Attributes')
ax.set_xlabel('GREAD  of performanc.'   )
ax.legend(loc='upper right')
plt.yticks(x ,cols_names)
ax.legend(bbox_to_anchor=(1,1))
ax.set_label(cols_names)
plt.show()

defenltly we witnce an increase in all atributes

sns.set_style("darkgrid")
plt.figure(figsize=(20,20), dpi= 180);
fig = px.histogram(data_frame=mo_df , x ="date" ,y = "overall_rating" ,nbins=32 
     , hover_data =mo_df ,title = "Mohammaed's Salah  cumulative overall rating",
       cumulative =True ,text_auto =True ,marginal ='box' );


fig.show();

<Figure size 3600x3600 with 0 Axes>

from his journey start as a professional we notice continuous incremental in his attributes but in corrugated way lets see every year

sns.set_style("darkgrid")
plt.figure(figsize=(20,20), dpi= 180);
fig = px.histogram(data_frame=mo_df , x ="date" ,y = "overall_rating" ,nbins=16 
     , hover_data =mo_df ,title = "Mohammaed's Salah  cumulative overall rating",
       cumulative =False ,text_auto =True ,marginal ='box' );


fig.show();

<Figure size 3600x3600 with 0 Axes>

First for the hist part :- the increase in mo's performance in 2013 then down in 2014 tyhen incres agian strongly

It is very important to note that 2016 only shows 2 matches in the first months, so we will ignore 2016 due to lack of overall score

From box plot , we notice that the third part of the overall performance ( 2q:3q )took only 10 months from September 2014 to July 2015, let's check it out in more detail

fig = px.line(mo_df, x='date', y='overall_rating' )
fig.add_scatter(x=mo_df['date'], y=mo_df['reactions'] ,name ='reactions', mode='lines',line=dict(color="gray"))
fig.add_scatter(x=mo_df['date'], y=mo_df['volleys'],name ="volleys", mode='lines',line=dict(color="black"))
fig.add_scatter(x=mo_df['date'], y=mo_df['balance'],name ="balance", mode='lines')
fig.add_scatter(x=mo_df['date'], y=mo_df['finishing'],name ="finishing", mode='lines')
fig.add_scatter(x=mo_df['date'], y=mo_df["sprint_speed"],name ="sprint_speed", mode='lines',line=dict(color="indigo"))
fig.add_scatter(x=mo_df['date'], y=mo_df['heading_accuracy'],name ="heading accuracy", mode='lines+markers')
fig.add_scatter(x=mo_df['date'], y=mo_df['ball_control'],name ="ball control", mode='lines',line=dict(color="red"))
fig.add_scatter(x=mo_df['date'], y=mo_df['acceleration'],name ="acceleration", mode='lines')
fig.add_hline(y=1, annotation_position="bottom right")
fig.update_layout(legend=dict(orientation="v",yanchor="middle", y=.5,xanchor="left",x=1) , title='<b>Mohamed\'s Salah performance during the period from 2008 to 2016</b>', xaxis_title='<b>DATE</b>',  yaxis_title='<b>The scale of performance</b>')
fig.update_layout(yaxis_range=[40,100])
 
# Add images
fig.update_layout(images=[dict(source="images.jpg",xref="paper", yref="paper",x=0.2,  y=.7,sizex=.37, sizey=.8,xanchor="left", yanchor="bottom",
sizing= "contain",opacity= 0.7,visible = True,layer= "below")],hovermode="x unified",autosize=True,
font=dict(family="Helvetica, Bold", size=14,color="Black"))
# Set templates
fig.update_layout(template = "plotly_white")
fig.add_vrect(x0="2013-04-01", x1="2015-4-30",annotation_text="<b>the amazing explosion of performance</b>", fillcolor="red", opacity=0.2, line_width=1) 
ax.legend(bbox_to_anchor=(3,3))
fig.show()

Finishing attacks. The chart , clearly shows that Mohamed Salah’s ability to end the attacks has developed when he was loaned to Italian Fiorentina in the period between January and May 2015, so his abilities to end the attacks during the Chelsea period were (68%), but it progressed in Fiorentina until it reached (76%). ) on (April 24, 2015), which is his last period with the Italian team.

Dealing with volleys. Air games in particular witnessed the most prominent development in Salah’s abilities during his loan to Fiorentina. When he arrived at Chelsea, his capabilities were limited to a maximum potential that reached (48%), but during his time with the Fiorentina team, he reached (68%) , which reflects a development Notable in Salah's handling of aerial games.

Ball_control. Although Salah reached high numbers in the ability to control the eight-year period represented by the statistic, during the year he spent at Chelsea, his ability to control the ball decreased to (81%), but it returned to rise during his loan period to Fiorentina to reach in the last His period with the Italian team (April 24, 2015) reached (85%). reactions.

Salah's reaction with Fiorentina improved than it was before, so his numbers regarding this were at the end of his period at Chelsea (69%), but at the end of his period with Fiorentina (April 24, 2015) to reach (74%).

Balance. Mohamed Salah's balancing abilities improved quickly with Fiorentina and within a short period, when he arrived at the Italian club, his numbers were at their best regarding that (80%), he succeeded in reaching (84%) and he did not succeed for another

---

Second question

To investigate the second question we will use theTeam_Attributes_df

merge player and player_att to creat a new df contains the player names and attributes

Player_Att_numric=Player_Attributes.iloc[:,np.r_[2,3,4,5, 9 :42] ]
merged_plyerName_plyerAtt = players.merge(Player_Att_numric, on="player_api_id",suffixes =("_name","_att"))

select Spanish teams team_df

spanish_teams_df = Team_df.loc[(Team_df['id'] >= 43035) & (Team_df['id'] <=  48358 )]

check for null values for both DF

spanish_teams_df.isnull().values.any()

False

# we already droped buildUpPlayDribbling col by drop_forty_null_column

Team_Attributes_df.isnull().values.any() 

False

Team_Attributes_df.duplicated().sum()

0

spanish_teams_df.duplicated().sum()

0

Great no null or duplicated values

names_of_spanish_teams_df =spanish_teams_df.iloc[: , np.r_ [1,3 ]]
names_of_spanish_teams_df.isnull().values.any() 

False

inner join between Spanish teams names and attributes

spanish_att_df = names_of_spanish_teams_df.merge(Team_Attributes_df, on="team_api_id",suffixes =("_am","_att"))

let's have a look in Spanish teams attributes we just selected from the team_df

spanish_att_df.head(3)

	team_api_id	team_long_name	id	team_fifa_api_id	date	buildUpPlaySpeed	buildUpPlaySpeedClass	buildUpPlayDribblingClass	buildUpPlayPassing	buildUpPlayPassingClass	buildUpPlayPositioningClass	chanceCreationPassing	chanceCreationPassingClass	chanceCreationCrossing	chanceCreationCrossingClass	chanceCreationShooting	chanceCreationShootingClass	chanceCreationPositioningClass	defencePressure	defencePressureClass	defenceAggression	defenceAggressionClass	defenceTeamWidth	defenceTeamWidthClass	defenceDefenderLineClass
0	10267	Valencia CF	1308	461	2010-02-22 00:00:00	30	Slow	Little	30	Short	Free Form	55	Normal	60	Normal	70	Lots	Organised	55	Medium	60	Press	60	Normal	Offside Trap
1	10267	Valencia CF	1309	461	2011-02-22 00:00:00	41	Balanced	Little	40	Mixed	Organised	56	Normal	55	Normal	64	Normal	Free Form	40	Medium	48	Press	54	Normal	Cover
2	10267	Valencia CF	1310	461	2012-02-22 00:00:00	20	Slow	Little	35	Mixed	Organised	38	Normal	56	Normal	55	Normal	Organised	51	Medium	38	Press	59	Normal	Cover

spanish_att_df.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 187 entries, 0 to 186
Data columns (total 25 columns):
 #   Column                          Non-Null Count  Dtype 
---  ------                          --------------  ----- 
 0   team_api_id                     187 non-null    int64 
 1   team_long_name                  187 non-null    object
 2   id                              187 non-null    int64 
 3   team_fifa_api_id                187 non-null    int64 
 4   date                            187 non-null    object
 5   buildUpPlaySpeed                187 non-null    int64 
 6   buildUpPlaySpeedClass           187 non-null    object
 7   buildUpPlayDribblingClass       187 non-null    object
 8   buildUpPlayPassing              187 non-null    int64 
 9   buildUpPlayPassingClass         187 non-null    object
 10  buildUpPlayPositioningClass     187 non-null    object
 11  chanceCreationPassing           187 non-null    int64 
 12  chanceCreationPassingClass      187 non-null    object
 13  chanceCreationCrossing          187 non-null    int64 
 14  chanceCreationCrossingClass     187 non-null    object
 15  chanceCreationShooting          187 non-null    int64 
 16  chanceCreationShootingClass     187 non-null    object
 17  chanceCreationPositioningClass  187 non-null    object
 18  defencePressure                 187 non-null    int64 
 19  defencePressureClass            187 non-null    object
 20  defenceAggression               187 non-null    int64 
 21  defenceAggressionClass          187 non-null    object
 22  defenceTeamWidth                187 non-null    int64 
 23  defenceTeamWidthClass           187 non-null    object
 24  defenceDefenderLineClass        187 non-null    object
dtypes: int64(11), object(14)
memory usage: 38.0+ KB

we will select the numeric attributes to explor team"s attributes

spanish_numric_att_list = ["buildUpPlaySpeed","buildUpPlayPassing","chanceCreationPassing","chanceCreationCrossing","chanceCreationShooting","defencePressure","defenceAggression","defenceTeamWidth"]
spanish_att_df[spanish_numric_att_list].head(2)

	buildUpPlaySpeed	buildUpPlayPassing	chanceCreationPassing	chanceCreationCrossing	chanceCreationShooting	defencePressure	defenceAggression	defenceTeamWidth
0	30	30	55	60	70	55	60	60
1	41	40	56	55	64	40	48	54

spanish_att_num_df =spanish_att_df[spanish_numric_att_list] 


spanish_att_num_df.describe()

	buildUpPlaySpeed	buildUpPlayPassing	chanceCreationPassing	chanceCreationCrossing	chanceCreationShooting	defencePressure	defenceAggression	defenceTeamWidth
count	187.000000	187.000000	187.000000	187.000000	187.000000	187.000000	187.000000	187.000000
mean	47.368984	45.839572	53.171123	53.716578	54.941176	47.828877	48.566845	55.850267
std	11.169645	10.997357	10.429622	10.835520	10.259628	9.231935	8.681929	8.793878
min	20.000000	23.000000	30.000000	24.000000	23.000000	30.000000	30.000000	30.000000
25%	40.000000	37.000000	45.000000	45.000000	50.000000	40.000000	41.500000	50.000000
50%	48.000000	44.000000	54.000000	54.000000	54.000000	49.000000	49.000000	56.000000
75%	55.000000	54.000000	59.000000	60.000000	60.000000	52.000000	53.000000	62.000000
max	71.000000	73.000000	73.000000	78.000000	79.000000	70.000000	70.000000	70.000000

by check mean min max looks like everything normal

Well, let's explore the offensive attributes of the teams

plt.figure(figsize=(6,6), dpi= 180) 
fig, ax = plt.subplots();
spanish_att_df.chanceCreationPassing.hist(bins=9);
ax.set_title("Spinsh teams chance Creation Passing")
ax.set_ylabel('frequency')
ax.set_xlabel('chance Creation Passing grade');

<Figure size 1080x1080 with 0 Axes>

plt.figure(figsize=(6,6), dpi= 180) 
fig, ax = plt.subplots();
spanish_att_df.chanceCreationShooting.hist(bins=5);
ax.set_title("Spinsh teams chance Creation Shooting")
ax.set_ylabel('frequency')
ax.set_xlabel('chance Creation Shooting grade');

<Figure size 1080x1080 with 0 Axes>

from the three charts above it's clear that this attribute very strong attributes

Well, let's explore the defensive attributes of the teams

spanish_att_df.defenceTeamWidth

0      60
1      54
2      59
3      59
4      59
       ..
182    67
183    57
184    57
185    57
186    57
Name: defenceTeamWidth, Length: 187, dtype: int64

plt.figure(figsize=(6,6), dpi= 180) 
fig, ax = plt.subplots();
spanish_att_df.defencePressure.hist(bins=9);
ax.set_title("Spinsh teams defence Pressure")
ax.set_ylabel('frequency')
ax.set_xlabel('defence Pressure grade');

<Figure size 1080x1080 with 0 Axes>

plt.figure(figsize=(6,6), dpi= 180) 
fig, ax = plt.subplots();
spanish_att_df.defenceAggression.hist(bins=9);
ax.set_title("Spinsh teams defence Aggression")
ax.set_ylabel('frequency')
ax.set_xlabel('defence Aggressio grade');

<Figure size 1080x1080 with 0 Axes>

plt.figure(figsize=(6,6), dpi= 180) 
fig, ax = plt.subplots();
spanish_att_df.defenceAggression.hist(bins=9);
ax.set_title("Spinsh teams defence Aggression")
ax.set_ylabel('frequency')
ax.set_xlabel('defence Aggression grade');

<Figure size 1080x1080 with 0 Axes>

it's clear that the grades for defense are less than offensive ones which represent the offensive style of the Spanish league

let's find out the top strong Spanish teams

grouping by team name all his numeric attributes and sum them

total_df=spanish_att_df.groupby("team_long_name")[ spanish_numric_att_list].sum()
total_df.head(1)
 

	buildUpPlaySpeed	buildUpPlayPassing	chanceCreationPassing	chanceCreationCrossing	chanceCreationShooting	defencePressure	defenceAggression	defenceTeamWidth
team_long_name
Athletic Club de Bilbao	310	288	280	421	281	238	374	341

total of all summation of attributes sorted descendingly

total_df["sum"] =total_df.sum(axis=1)
total_df["sum"]
#total_df["sum"].sort_values(ascending= False)

team_long_name
Athletic Club de Bilbao      2533
Atlético Madrid              2624
CA Osasuna                   2560
CD Numancia                  2403
CD Tenerife                  1672
Córdoba CF                   2174
Elche CF                     2485
FC Barcelona                 2320
Getafe CF                    2189
Granada CF                   2221
Hércules Club de Fútbol      2100
Levante UD                   2515
Málaga CF                    2499
RC Celta de Vigo             2479
RC Deportivo de La Coruña    2392
RC Recreativo                2190
RCD Espanyol                 2422
RCD Mallorca                 2513
Racing Santander             2168
Rayo Vallecano               2356
Real Betis Balompié          2464
Real Madrid CF               2684
Real Sociedad                2398
Real Sporting de Gijón       2591
Real Valladolid              2494
Real Zaragoza                2240
SD Eibar                      857
Sevilla FC                   2646
UD Almería                   2505
UD Las Palmas                2364
Valencia CF                  2247
Villarreal CF                2295
Xerez Club Deportivo         1562
Name: sum, dtype: int64

top ten Spanish teams

TOp_TEN =total_df["sum"].sort_values(ascending= False).head(10)
TOp_TEN 

team_long_name
Real Madrid CF             2684
Sevilla FC                 2646
Atlético Madrid            2624
Real Sporting de Gijón     2591
CA Osasuna                 2560
Athletic Club de Bilbao    2533
Levante UD                 2515
RCD Mallorca               2513
UD Almería                 2505
Málaga CF                  2499
Name: sum, dtype: int64

plt.rcParams.update({'font.size': 19})

plt.figure(figsize=(15,12), dpi= 180) 
TOp_TEN .plot(kind='bar', title='Spanish Top ten in total attributes 2008:2016', ylabel='total greads over the period',
         xlabel='teams names', figsize=(6, 5));

Something is wrong here where FC Barcelona 3-time champion from 2008 to 2016

total_df["sum"].sort_values(ascending= False).head(21)

team_long_name
Real Madrid CF               2684
Sevilla FC                   2646
Atlético Madrid              2624
Real Sporting de Gijón       2591
CA Osasuna                   2560
Athletic Club de Bilbao      2533
Levante UD                   2515
RCD Mallorca                 2513
UD Almería                   2505
Málaga CF                    2499
Real Valladolid              2494
Elche CF                     2485
RC Celta de Vigo             2479
Real Betis Balompié          2464
RCD Espanyol                 2422
CD Numancia                  2403
Real Sociedad                2398
RC Deportivo de La Coruña    2392
UD Las Palmas                2364
Rayo Vallecano               2356
FC Barcelona                 2320
Name: sum, dtype: int64

The result is not even near to the fact that FC Barcelona is the champion of three-time in the exact period so let's modify our range to focus on the attributes relate to any strong team the offensive part and the strength of the formation attacks and shooting :-

* chanceCreationShooting

* chanceCreationPassing

* buildUpPlaySpeed

* buildUpPlayPassing

spanish_numric_att_limit_list = ["buildUpPlaySpeed","buildUpPlayPassing","chanceCreationPassing","chanceCreationShooting"]
limted_total_df=spanish_att_df.groupby("team_long_name")[ spanish_numric_att_limit_list].sum()
limted_total_df

	buildUpPlaySpeed	buildUpPlayPassing	chanceCreationPassing	chanceCreationShooting
team_long_name
Athletic Club de Bilbao	310	288	280	281
Atlético Madrid	319	287	388	360
CA Osasuna	238	297	321	339
CD Numancia	302	273	304	336
CD Tenerife	198	195	225	233
Córdoba CF	242	278	250	307
Elche CF	289	332	309	344
FC Barcelona	215	204	271	318
Getafe CF	207	238	302	332
Granada CF	262	222	326	262
Hércules Club de Fútbol	253	247	283	309
Levante UD	327	340	350	349
Málaga CF	291	253	348	331
RC Celta de Vigo	292	298	316	336
RC Deportivo de La Coruña	307	257	239	385
RC Recreativo	255	303	268	220
RCD Espanyol	319	271	350	284
RCD Mallorca	337	320	317	315
Racing Santander	236	246	318	257
Rayo Vallecano	253	261	255	319
Real Betis Balompié	314	244	334	339
Real Madrid CF	304	232	405	414
Real Sociedad	241	231	338	359
Real Sporting de Gijón	354	287	398	305
Real Valladolid	317	322	302	302
Real Zaragoza	235	201	307	340
SD Eibar	124	104	119	116
Sevilla FC	342	336	350	315
UD Almería	300	363	330	299
UD Las Palmas	276	262	324	341
Valencia CF	176	210	262	357
Villarreal CF	233	194	273	369
Xerez Club Deportivo	190	176	181	201

total of all summation of limet list of attributes sorted descendingly

limted_total_df["sum"] =limted_total_df.sum(axis=1)

limted_total_df["sum"].sort_values(ascending= False)

team_long_name
Levante UD                   1366
Real Madrid CF               1355
Atlético Madrid              1354
Real Sporting de Gijón       1344
Sevilla FC                   1343
UD Almería                   1292
RCD Mallorca                 1289
Elche CF                     1274
Real Valladolid              1243
RC Celta de Vigo             1242
Real Betis Balompié          1231
RCD Espanyol                 1224
Málaga CF                    1223
CD Numancia                  1215
UD Las Palmas                1203
CA Osasuna                   1195
RC Deportivo de La Coruña    1188
Real Sociedad                1169
Athletic Club de Bilbao      1159
Hércules Club de Fútbol      1092
Rayo Vallecano               1088
Real Zaragoza                1083
Getafe CF                    1079
Córdoba CF                   1077
Granada CF                   1072
Villarreal CF                1069
Racing Santander             1057
RC Recreativo                1046
FC Barcelona                 1008
Valencia CF                  1005
CD Tenerife                   851
Xerez Club Deportivo          748
SD Eibar                      463
Name: sum, dtype: int64

top ten Spanish teams for most important attributes related to offensive style in playing soccer (football)

TOp_limt_TEN =limted_total_df["sum"].sort_values(ascending= False).head(10)
TOp_limt_TEN

team_long_name
Levante UD                1366
Real Madrid CF            1355
Atlético Madrid           1354
Real Sporting de Gijón    1344
Sevilla FC                1343
UD Almería                1292
RCD Mallorca              1289
Elche CF                  1274
Real Valladolid           1243
RC Celta de Vigo          1242
Name: sum, dtype: int64

plt.rcParams.update({'font.size': 19})

plt.figure(figsize=(15,12), dpi= 180) 
TOp_limt_TEN.plot(kind='barh', title='Spanish Top ten Limited List in total attributes 2008:2016', ylabel='total greads over the period',
         xlabel='teams names', figsize=(6, 5));

No trace of Barcelona in the top ten agian !!!

limted_total_df["sum"].sort_values(ascending= False).head(29)

team_long_name
Levante UD                   1366
Real Madrid CF               1355
Atlético Madrid              1354
Real Sporting de Gijón       1344
Sevilla FC                   1343
UD Almería                   1292
RCD Mallorca                 1289
Elche CF                     1274
Real Valladolid              1243
RC Celta de Vigo             1242
Real Betis Balompié          1231
RCD Espanyol                 1224
Málaga CF                    1223
CD Numancia                  1215
UD Las Palmas                1203
CA Osasuna                   1195
RC Deportivo de La Coruña    1188
Real Sociedad                1169
Athletic Club de Bilbao      1159
Hércules Club de Fútbol      1092
Rayo Vallecano               1088
Real Zaragoza                1083
Getafe CF                    1079
Córdoba CF                   1077
Granada CF                   1072
Villarreal CF                1069
Racing Santander             1057
RC Recreativo                1046
FC Barcelona                 1008
Name: sum, dtype: int64

limted_total_df.describe()

	buildUpPlaySpeed	buildUpPlayPassing	chanceCreationPassing	chanceCreationShooting	sum
count	33.000000	33.000000	33.000000	33.000000	33.000000
mean	268.424242	259.757576	301.303030	311.333333	1140.818182
std	53.680205	54.845254	58.592067	57.570971	187.168449
min	124.000000	104.000000	119.000000	116.000000	463.000000
25%	236.000000	231.000000	271.000000	299.000000	1072.000000
50%	276.000000	261.000000	309.000000	319.000000	1188.000000
75%	310.000000	297.000000	334.000000	341.000000	1243.000000
max	354.000000	363.000000	405.000000	414.000000	1366.000000

so let's extract the mean and median for the 4 atrributes

buildUpPlaySpeed mean = 268 < median = 276
buildUpPlayPassing mean = 260 < median = 261
chanceCreationPassing mean = 301 < median = 309
chanceCreationShooting mean = 311 < median = 319

from above we have left skew which in general the Spanish teams have strong grades in these attributes

# total of all summation of selected as representative attributes

limted_total_df["sum"] =limted_total_df.sum(axis=1)
limted_total_sum_df=limted_total_df.iloc[ :,:].sort_values(by="sum",ascending= False)
limted_total_sum_df.head(29).style.bar()

	buildUpPlaySpeed	buildUpPlayPassing	chanceCreationPassing	chanceCreationShooting	sum
team_long_name
Levante UD	327	340	350	349	2732
Real Madrid CF	304	232	405	414	2710
Atlético Madrid	319	287	388	360	2708
Real Sporting de Gijón	354	287	398	305	2688
Sevilla FC	342	336	350	315	2686
UD Almería	300	363	330	299	2584
RCD Mallorca	337	320	317	315	2578
Elche CF	289	332	309	344	2548
Real Valladolid	317	322	302	302	2486
RC Celta de Vigo	292	298	316	336	2484
Real Betis Balompié	314	244	334	339	2462
RCD Espanyol	319	271	350	284	2448
Málaga CF	291	253	348	331	2446
CD Numancia	302	273	304	336	2430
UD Las Palmas	276	262	324	341	2406
CA Osasuna	238	297	321	339	2390
RC Deportivo de La Coruña	307	257	239	385	2376
Real Sociedad	241	231	338	359	2338
Athletic Club de Bilbao	310	288	280	281	2318
Hércules Club de Fútbol	253	247	283	309	2184
Rayo Vallecano	253	261	255	319	2176
Real Zaragoza	235	201	307	340	2166
Getafe CF	207	238	302	332	2158
Córdoba CF	242	278	250	307	2154
Granada CF	262	222	326	262	2144
Villarreal CF	233	194	273	369	2138
Racing Santander	236	246	318	257	2114
RC Recreativo	255	303	268	220	2092
FC Barcelona	215	204	271	318	2016

from a glimpse to the result again even with limiting our exploration to the most related attributes still this result as a total or as attributes individually compared to the other teams is not even close to representing the actually, Barcelona results are almost near the median in most

Third Question

---

we already dropped 17 col cause they had 40% null values from Match DataFrame

select the columns we interested to explore from Match df

match_df=match_df.iloc[: , np.r_ [2:11 ]]

spanish_match_df = match_df[match_df["league_id"]==21518]
spanish_match_df.head(1) 

	league_id	season	stage	date	match_api_id	home_team_api_id	away_team_api_id	home_team_goal	away_team_goal
21517	21518	2008/2009	1	2008-08-30 00:00:00	530023	10267	8661	3	0

spanish_match_df.shape

(3040, 9)

spanish_match_df.isnull().values.any()

False

spanish_match_df.duplicated().sum()

0

great no null or duplicated values

creat sepreat DataFrame copy for home points

home_names_of_spanish_teams_df=names_of_spanish_teams_df.copy()

change name of team api_id to use it in mereg

home_names_of_spanish_teams_df.rename(columns={'team_api_id': 'home_team_api_id'}, inplace=True)
home_names_of_spanish_teams_df.head(1)

	home_team_api_id	team_long_name
251	10267	Valencia CF

calculate the point/s for home team per match

spanish_match_df.loc[:-2,"home_point"] = 0

spanish_match_df["home_point"]= np.select(condlist=[spanish_match_df["home_team_goal"] > spanish_match_df["away_team_goal"],
                                         spanish_match_df["home_team_goal"] == spanish_match_df["away_team_goal"]],
                                                                                   choicelist=[3,1],default=0) 
spanish_match_df

	league_id	season	stage	date	match_api_id	home_team_api_id	away_team_api_id	home_team_goal	away_team_goal	home_point
21517	21518	2008/2009	1	2008-08-30 00:00:00	530023	10267	8661	3	0	3
21518	21518	2008/2009	1	2008-08-31 00:00:00	530084	8371	10205	1	1	1
21519	21518	2008/2009	1	2008-08-31 00:00:00	530085	9783	8633	2	1	3
21520	21518	2008/2009	1	2008-08-31 00:00:00	530086	8388	8634	1	0	3
21521	21518	2008/2009	1	2008-08-31 00:00:00	530087	8696	8302	1	1	1
...	...	...	...	...	...	...	...	...	...	...
24552	21518	2015/2016	9	2015-10-25 00:00:00	2030167	9906	10267	2	1	3
24553	21518	2015/2016	9	2015-10-24 00:00:00	2030168	9864	9783	2	0	3
24554	21518	2015/2016	9	2015-10-26 00:00:00	2030169	8315	9869	3	0	3
24555	21518	2015/2016	9	2015-10-24 00:00:00	2030170	7878	8603	1	1	1
24556	21518	2015/2016	9	2015-10-23 00:00:00	2030171	8370	8558	3	0	3

3040 rows × 10 columns

select the columns relate to home to exploer

home_spanish_match_df = spanish_match_df.iloc[: , np.r_ [1,3,5,7,9 ]]   
home_spanish_match_df.head(1)

	season	date	home_team_api_id	home_team_goal	home_point
21517	2008/2009	2008-08-30 00:00:00	10267	3	3

merge "inner join" for home Results

result_home_spanish_df =home_names_of_spanish_teams_df.merge(home_spanish_match_df, on="home_team_api_id")
# parssing the date
result_home_spanish_df["date"]= pd.to_datetime(result_home_spanish_df["date"])

spanish_match_df.head(2)

	league_id	season	stage	date	match_api_id	home_team_api_id	away_team_api_id	home_team_goal	away_team_goal	home_point
21517	21518	2008/2009	1	2008-08-30 00:00:00	530023	10267	8661	3	0	3
21518	21518	2008/2009	1	2008-08-31 00:00:00	530084	8371	10205	1	1	1

Let's calculate the home and away goals for each season

goals_per_season =spanish_match_df.groupby("season")[ ["home_team_goal","away_team_goal"]].sum()

goals_per_season

	home_team_goal	away_team_goal
season
2008/2009	631	470
2009/2010	608	423
2010/2011	622	420
2011/2012	638	412
2012/2013	641	450
2013/2014	620	425
2014/2015	584	425
2015/2016	615	428

fig, ax = plt.subplots(figsize=(15,6), dpi= 180)
goals_per_season["home_team_goal"].plot(ax=ax)
goals_per_season["away_team_goal"].plot(ax=ax) 
plt.title("The spanish home vs. away goals   2008:2016")
plt.xlabel("Season")
plt.ylabel("POINTS")
plt.show()

result_home_spanish_df.head(2)

	home_team_api_id	team_long_name	season	date	home_team_goal	home_point
0	10267	Valencia CF	2008/2009	2008-08-30	3	3
1	10267	Valencia CF	2008/2009	2008-11-15	2	0

let's calulate the home golas for each team from2008:2016

goals_home_per_teams =result_home_spanish_df.groupby("team_long_name")["home_team_goal"].sum().sort_values(ascending= False).head(10)

goals_home_per_teams

team_long_name
Real Madrid CF             505
FC Barcelona               495
Atlético Madrid            321
Valencia CF                299
Sevilla FC                 285
Athletic Club de Bilbao    250
Málaga CF                  222
Villarreal CF              218
RCD Espanyol               205
Getafe CF                  202
Name: home_team_goal, dtype: int64

goals_home_per_teams.describe()

count     10.000000
mean     300.200000
std      112.868458
min      202.000000
25%      219.000000
50%      267.500000
75%      315.500000
max      505.000000
Name: home_team_goal, dtype: float64

title_s ='Spanish Top ten as home goals 2008:2016'
ylab = 'total goals over the period'

goals_home_per_teams.plot(kind='bar', title =title_s, ylabel= ylab,xlabel='Home Team', figsize=(6, 5));

we notice Atlético Madrid in third place but far away from big 2

The list of top 10 in Spanish league from 2008:20016 according home point

#top ten team home  points 

Hr = result_home_spanish_df.groupby(["team_long_name"])["home_point"].sum().sort_values(ascending= False )#.unstack().plot(kind="bar") 
Hr.iloc[:11]

team_long_name
FC Barcelona               405
Real Madrid CF             398
Atlético Madrid            333
Valencia CF                304
Sevilla FC                 300
Athletic Club de Bilbao    271
Villarreal CF              261
Málaga CF                  241
RCD Espanyol               241
Getafe CF                  225
Real Sociedad              200
Name: home_point, dtype: int32

Hr.describe()

count     33.000000
mean     156.333333
std      112.506574
min        9.000000
25%       47.000000
50%      133.000000
75%      241.000000
max      405.000000
Name: home_point, dtype: float64

from describe() we notice ateam have only 9 points at home in 8 sessions

Hr_di = result_home_spanish_df.groupby(["team_long_name"])["home_point"].sum().sort_values(ascending= False )
Hr_di.iloc[-1:]

team_long_name
Córdoba CF    9
Name: home_point, dtype: int32

it's hard to believe that the last team point for 8 sessions in his home only 9 points!!!! Córdoba CF 9

Hr_di=Hr_di.iloc[:11].sort_values(ascending= False ); 

title_s ='Spanish Top ten as home points 2008:2016'
ylab = 'total points over the period'

Hr_di.plot(kind='bar', title =title_s, ylabel= ylab,xlabel='Home Team', figsize=(6, 5))

<AxesSubplot:title={'center':'Spanish Top ten as home points 2008:2016'}, xlabel='Home Team', ylabel='total points over the period'>

we notice Atlético Madrid in third place agian !

creat sepreat DataFrame copy for away points

creat sepreat copy for away name

away_names_of_spanish_teams_df =names_of_spanish_teams_df.copy()

change name of team api_id to use it in mereg

away_names_of_spanish_teams_df.rename(columns={'team_api_id': 'away_team_api_id'}, inplace=True)
away_names_of_spanish_teams_df.head(1)

	away_team_api_id	team_long_name
251	10267	Valencia CF

calculate the point/s for away team per match

spanish_match_df.loc[:-2,"away_point"] = 0
spanish_match_df["away_point"]= np.select(condlist=[spanish_match_df["home_team_goal"] < spanish_match_df["away_team_goal"],
                                         spanish_match_df["home_team_goal"] == spanish_match_df["away_team_goal"]],choicelist=[3,1],default=0) 

select the columns relate to away to exploer

away_spanish_match_df = spanish_match_df.iloc[: , np.r_ [1,3,6,8,10 ]]
away_spanish_match_df.head(1)

	season	date	away_team_api_id	away_team_goal	away_point
21517	2008/2009	2008-08-30 00:00:00	8661	0	0

merge "inner join" for away Results

 result_away_spanish_df = away_names_of_spanish_teams_df.merge(away_spanish_match_df, on="away_team_api_id")
# parssing the date
result_away_spanish_df["date"]= pd.to_datetime(result_away_spanish_df["date"])

calculate the total points fro all teams in the period from 2008:2016

Ar=result_away_spanish_df.groupby(["team_long_name"] )["away_point"].sum().sort_values(ascending= False) 

top ten teams according to points away

Ar =Ar.iloc[:11].sort_values(ascending= False ); 

title_s ='Spanish Top ten as away points 2008:2016'
ylab = 'total points over the period'

Ar.plot(kind='bar', title =title_s, ylabel= ylab,xlabel='away Team', figsize=(6, 5))

<AxesSubplot:title={'center':'Spanish Top ten as away points 2008:2016'}, xlabel='away Team', ylabel='total points over the period'>

top ten teams according to goals away

and Atlético Madrid in third place agian !

goals_away_per_teams=result_away_spanish_df.groupby(["team_long_name"] )["away_team_goal"].sum().sort_values(ascending= False) 
goals_away_per_teams= goals_away_per_teams.iloc[:11].sort_values(ascending= False );

title_s ='Spanish Top ten as away goals 2008:2016'
ylab = 'total goals over the period'

goals_away_per_teams.plot(kind='bar', title =title_s, ylabel= ylab,xlabel='away Team', figsize=(6, 5));

definitely something desiver to investigate in depith hear . we notice Atlético Madrid in third place agian ! as pattern but have alook to stats

Ar.describe()

count     11.000000
mean     190.545455
std       77.266246
min      111.000000
25%      137.000000
50%      166.000000
75%      211.500000
max      340.000000
Name: away_point, dtype: float64

wow, we notice a team has only 5 points for 8 S away

we wonder who is it?

goals_away_per_teams.describe()

count     11.000000
mean     195.000000
std       78.467828
min      130.000000
25%      144.500000
50%      165.000000
75%      198.000000
max      354.000000
Name: away_team_goal, dtype: float64

Ar=result_away_spanish_df.groupby(["team_long_name"] )["away_point"].sum().sort_values(ascending= True) 
Ar.iloc[:1]

team_long_name
CD Numancia    5
Name: away_point, dtype: int32

CD Numancia 5

lets investigate the top three Spanish teams

top_three = ["Atlético Madrid","Real Madrid CF", "FC Barcelona"]
 # select only the 3 teams results from home  goal & point 
home_top_3 =result_home_spanish_df.query(f'team_long_name in {top_three}') 
home_top_3.head(1)

	home_team_api_id	team_long_name	season	date	home_team_goal	home_point
608	8633	Real Madrid CF	2008/2009	2008-11-08	4	3

summiton of the home points grouping by season then name for each team

Hr_home_top_3 = home_top_3.groupby(["season","team_long_name"])["home_point"].sum() 
Hr_home_top_3

season     team_long_name 
2008/2009  Atlético Madrid    40
           FC Barcelona       45
           Real Madrid CF     44
2009/2010  Atlético Madrid    36
           FC Barcelona       55
           Real Madrid CF     54
2010/2011  Atlético Madrid    33
           FC Barcelona       50
           Real Madrid CF     49
2011/2012  Atlético Madrid    38
           FC Barcelona       52
           Real Madrid CF     50
2012/2013  Atlético Madrid    44
           FC Barcelona       55
           Real Madrid CF     53
2013/2014  Atlético Madrid    49
           FC Barcelona       50
           Real Madrid CF     49
2014/2015  Atlético Madrid    45
           FC Barcelona       49
           Real Madrid CF     50
2015/2016  Atlético Madrid    48
           FC Barcelona       49
           Real Madrid CF     49
Name: home_point, dtype: int32

select only the 3 teams results from away goal & point

top_three = ["Atlético Madrid","Real Madrid CF", "FC Barcelona"]

 
away_top_3 =result_away_spanish_df.query(f'team_long_name in {top_three}') 
away_top_3.head(1)

	away_team_api_id	team_long_name	season	date	away_team_goal	away_point
608	8633	Real Madrid CF	2008/2009	2008-08-31	1	0

sum of the away points per season then name of each team

Ar_away_top_3 = away_top_3.groupby(["season","team_long_name"])["away_point"].sum()
Ar_away_top_3

season     team_long_name 
2008/2009  Atlético Madrid    27
           FC Barcelona       42
           Real Madrid CF     34
2009/2010  Atlético Madrid    11
           FC Barcelona       44
           Real Madrid CF     42
2010/2011  Atlético Madrid    25
           FC Barcelona       46
           Real Madrid CF     43
2011/2012  Atlético Madrid    18
           FC Barcelona       39
           Real Madrid CF     50
2012/2013  Atlético Madrid    32
           FC Barcelona       45
           Real Madrid CF     32
2013/2014  Atlético Madrid    41
           FC Barcelona       37
           Real Madrid CF     38
2014/2015  Atlético Madrid    33
           FC Barcelona       45
           Real Madrid CF     42
2015/2016  Atlético Madrid    40
           FC Barcelona       42
           Real Madrid CF     41
Name: away_point, dtype: int32

total_home_top_3 = home_top_3.iloc[: , np.r_ [1,2,3,5 ]]
total_home_top_3_1=total_home_top_3.rename({'home_point': 'points'}, axis=1)


total_away_top_3 =away_top_3.iloc[: , np.r_ [1,2,3,5 ]] 
total_away_top_3_2 =total_away_top_3.rename({'away_point': 'points'}, axis=1)

concat the two df hom and away for top three

# concat the two df hom and away for top three 
total_top_3_all =pd.concat([total_home_top_3_1, total_away_top_3_2], axis=0)

total_top_3_all 

	team_long_name	season	date	points
608	Real Madrid CF	2008/2009	2008-11-08	3
609	Real Madrid CF	2008/2009	2008-11-22	3
610	Real Madrid CF	2008/2009	2008-12-07	0
611	Real Madrid CF	2008/2009	2008-12-20	3
612	Real Madrid CF	2008/2009	2009-01-04	3
...	...	...	...	...
2142	Atlético Madrid	2015/2016	2016-04-20	3
2143	Atlético Madrid	2015/2016	2016-05-08	0
2144	Atlético Madrid	2015/2016	2015-09-19	3
2145	Atlético Madrid	2015/2016	2015-09-26	0
2146	Atlético Madrid	2015/2016	2015-10-18	3

912 rows × 4 columns

fig, ax = plt.subplots(figsize=(14,5), dpi= 180)
total_top_3_all.groupby(["season" ,"team_long_name" ])[ "points"].sum().unstack().plot(ax=ax)    
plt.title("The spanish top tree teams 2008:2016")
plt.xlabel("Seasons")
plt.ylabel("POINTS")
plt.show()
plt.tight_layout();

<Figure size 432x288 with 0 Axes>

remarkable journey from the 9th in 2009 to the champion in 2013 Atlético Madrid was the champion la Liga 2013/2014 Atlético Madrid 90 FC Barcelona 87 Real Madrid CF 87 Atlético repeated his achievement again and won the championship but after 2016 which out analysis scope

ttt_groubed=total_top_3_all.groupby(["team_long_name","season"])["points"].sum();

plt.rcParams.update({'font.size': 19})

ax =ttt_groubed.unstack().plot(kind ="barh" , figsize=(14,15));
ax.legend(bbox_to_anchor=(1,1))
plt.title("The spanish top tree teams 2008:2016")
plt.ylabel("teams")
plt.xlabel("POINTS")
plt.show()
plt.tight_layout();

<Figure size 432x288 with 0 Axes>

From the bottom bar of each team representing 2008 to the upper one, we see an increase in Atlético points compared to Real Madrid and Barcelona

Conclusions

While tracing Mohamed Salah's career in different clubs he played for, I witnessed a good rise in his performance in the period between 2012-2014. Then I found a decrease in his effort and then a greater rise within a short period, specifically the first months of 2015. Salah’s achievement and results changed so much during these periods, I found that the first period of rising was the period of his first professional career in Europe with the Swiss FC Basel, and it lasted for a year and a half, while the period of the greatest rise in his career during the years of research was the period of the first months of The year 2015, specifically between February and May 2015, was a period during which he loaned the English club, Chelsea, to Fiorentina.

the Spanish league main characteristics is an offensive style in chanceCreationShooting chanceCreationPassing buildUpPlaySpeed * buildUpPlayPassing as normal real Madrid shows a great grade attributes but the dataset failed to represent the same situation with the champion Barcelona

in investagte the third Question we face arepeted batterrn that It is clear to us that the claims that lack of competition in La Liga and constrained it only between the two big teams Real Madrid and Barcelona are inaccurate and that Atlético de Madrid has won the league title once and with a very good performance since 2008: 2016 from the ninth. to the top three. Many times and It's a strange result to find out that CD Numancia is 5 points away and Córdoba CF only has 9 points!!!! In all eight seasons**

Limitations

• ### It was noticed while tracking the table of team attributes of the Spanish teams in both the total numerical attributes of the league champion Barcelona team, and even with the modification of the search and limiting it only to the distinctive attributes of the top teams, I mean the offensive side and the formation of the attack, the results derived from the table did not match with the Barcelona team real results,I can't find the reason for the low grades for a champion 3 times in the exact period

• Despite noting the improvement and decline in Salah's performance across different clubs and his very big improvement at Fiorentina, we couldn't figure out the reason for this improvement. This cannot be known by looking at this data.

for curiosity i investigated the English league using the same table team attributes the results are not consistent again with real results. top teams in the bottom !! below is a print screen of the final result

acknowledges to

www.geeksforgeeks.org www.stackoverflow.com www.analyticsvidhya.com